5-bit Open on Hugging Face
Parameters
9.4B
Weights (est.)
≈ 7.1 GB
Mac memory (est.)
16 GB+
Size is estimated from the parameter count and quantisation. Memory allows a fifth extra for context plus 2 GB for the system; long contexts need more.
Run it
With mlx-vlm on an Apple silicon Mac:
pip install mlx-vlmpython -m mlx_vlm.generate --model mlx-community/Qwen3.5-9B-5bit --image photo.jpg --prompt "Describe this image"
Details
- Base model
- Qwen/Qwen3.5-9B-Base (quantized)
- Vendor
- Qwen
- Task
- Vision + text
- Quantisation
- 5-bit
- Licence
- apache-2.0
- Downloads
- 276 in the last 30 days · 7.3K all time · 0 likes
- Published
- 2026-03-02 · updated 2026-03-02 (8mo ago)
- Tags
- qwen3_5
Other MLX builds of Qwen/Qwen3.5-9B-Base
| Build | Quantisation | Weights | Mac memory | Downloads |
|---|---|---|---|---|
| Qwen3.5-9B-4bit | 4-bit | ≈ 6.0 GB | 16 GB+ | 25K |
| Qwen3.5-9B-5bit (this one) | 5-bit | ≈ 7.1 GB | 16 GB+ | 276 |
| Qwen3.5-9B-6bit | 6-bit | ≈ 8.2 GB | 16 GB+ | 1.2K |
| Qwen3.5-9B-8bit | 8-bit | ≈ 10.4 GB | 16 GB+ | 3.0K |
| Qwen3.5-9B-bf16 | bf16 | ≈ 18.8 GB | 24 GB+ | 1.7K |